CursorCore: Assist Programming through Aligning Anything

TL;DR

CursorCore introduces Assistant-Conversation, a unified framework to align $S$, $H$, $C$, and $U$ with $A$ for programming tasks, addressing limitations of traditional completion, insertion, and chat-based assistants. It pairs a new data-generation pipeline, Programming-Instruct, with the APEval benchmark to evaluate how well models leverage history, context, and instructions during coding. The authors fine-tune open-source LLMs into CursorCore variants, achieving state-of-the-art or near state-of-the-art results across multiple languages and benchmarks, and demonstrate effectiveness in both code editing and inline chat scenarios. This work suggests a practical path to more automated, history-aware programming assistants and paves the way for repository- and IDE-integrated development tools with broad applicability.

Abstract

Large language models have been successfully applied to programming assistance tasks, such as code completion, code insertion, and instructional code editing. However, these applications remain insufficiently automated and struggle to effectively integrate various types of information during the programming process, including coding history, current code, and user instructions. In this work, we propose a new conversational framework that comprehensively integrates these information sources, collect data to train our models and evaluate their performance. Firstly, to thoroughly evaluate how well models align with different types of information and the quality of their outputs, we introduce a new benchmark, APEval (Assist Programming Eval), to comprehensively assess the performance of models in programming assistance tasks. Then, for data collection, we develop a data generation pipeline, Programming-Instruct, which synthesizes training data from diverse sources, such as GitHub and online judge platforms. This pipeline can automatically generate various types of messages throughout the programming process. Finally, using this pipeline, we generate 219K samples, fine-tune multiple models, and develop the CursorCore series. We show that CursorCore outperforms other models of comparable size. This framework unifies applications such as inline chat and automated editing, contributes to the advancement of coding assistants. Code, models and data are freely available at https://github.com/TechxGenus/CursorCore.

Figures (24)

• Figure 1: Different forms of programming assistance. The common uses of current LLMs are shown on the left. Our framework is shown on the right.
• Figure 2: Examples of Assistant-Conversation from our training data. The top example demonstrates predicting the corresponding edits and explanations based on historical edits and the current code context. The bottom example demonstrates predictions based on the current code and user instructions.
• Figure 3: Samples from AI Programmer, Git Commit and Online Judge Submission.
• Figure 4: Data processing pipeline. The randomly selected time point is the third, data type is H and C.
• Figure 5: Distribution of programming language in the training data.